INDEX

Explanations

mentions of legal or political contexts like FISA and government intervention

oai_token-act-pair · gpt-3.5-turbo

references to significant actions or positions taken by authority figures or institutions

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.9.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Belfast

-0.84

 Adelaide

-0.80

 neighbouring

-0.78

 Newcastle

-0.75

 Britons

-0.74

 Cardiff

-0.73

 Nato

-0.71

 Rugby

-0.70

 Fifa

-0.70

 Calais

-0.70

POSITIVE LOGITS

 nonexistent

0.73

though

0.66

arro

0.66

uliffe

0.63

 malf

0.61

 basically

0.60

 equivalents

0.60

 crappy

0.59

farious

0.59

 disav

0.59

Activations Density 1.738%

mentions of legal or political contexts like FISA and government intervention

references to significant actions or positions taken by authority figures or institutions

No Comments

No Known Activations